Automated Carrot Harvesting Machine With YOLOv8 for Precision and Optimal Efficiency

Carrots are a key staple in Pakistan’s agriculture, yet harvesting practices remain predominantly manual, resulting in high labor costs, inefficiencies, and considerable postharvest losses. The current study presents the design and fabrication of a cost‐effective, intelligent carrot harvesting machine, modeled in SolidWorks and optimized for key operational parameters: claw belt speed of 4 m/s, roller speed of 1.2 m/s, and a taper angle of 26°, to maximize pick‐up efficiency and minimize crop damage. A YOLOv8‐based quality assessment model, trained on a region‐specific annotated dataset of local carrot varieties, was integrated for real‐time defect detection. The model achieved high accuracy (approximately 0.98), F 1‐score (approximately 0.95), and mAP@0.5 (approximately 0.94), ensuring the reliable sorting of high‐quality produce. Laboratory evaluations demonstrated significant performance gains over manual harvesting methods in terms of speed (3–5 acres/day vs. 0.2–0.5 acres/day), efficiency (80%–92%), and reduced physical strain. These findings support the adoption of mechanized harvesting aligned with precision agriculture to enhance productivity, safety, and sustainability.